Search Results

You are looking at 1 - 10 of 25 items for :

  • Author or Editor: David P. Baumhefner x
  • Refine by Access: All Content x
Clear All Modify Search
David P. Baumhefner

Abstract

A pilot study that evaluates the potential forecast skill of winter 10–30-day time-mean flow from a low-resolution (R15) climate simulation model is presented. The hypothesis tested is that low-resolution climate model forecasts might be as skillful as high-resolution numerical weather prediction model forecasts at extended-range timescales, if the low-frequency evolution is primarily a large-scale process and if the systematic error of the climate model is less detrimental than high-resolution forecast model error.

Eight forecast cases, each containing four ensemble members, are examined and compared to high-resolution forecasts discussed by Miyakoda et al. The systematic error of the climate model is examined and then used to reduce the forecast error in an a posteriors fashion. The operational utility of these climate model forecasts is also assessed.

The low-resolution climate model is quite successful in duplicating the skill of the high-resolution forecast model. If the forecast systematic component of error evaluated from the same eight cases is removed, the climate model forecasts improve in a comparable fashion to the high-resolution results. When information from the low-resolution climate simulation is used to estimate the forecast systematic error, the improvement in skill is less successful. These results show that a low-resolution climate model can be a viable tool for numerical extended-range forecasting and imply that large ensembles can be integrated for the same cost as higher-resolution model integrations.

Full access
David P. Baumhefner

Abstract

An attempt is made to investigate the effect of the tropics and the Southern Hemisphere on Northern Hemisphere mid-latitude numerical forecasting. Using the NCAR general circulation model and real atmospheric data for January 1958, several experiments were conducted in which a wall was inserted at various latitudes. These forecasts were compared with real data and a global, real-data forecast without a wall. Verification statistics were evaluated for comparison purposes. Several different boundary conditions at the wall were also examined.

Walls inserted in the model at or below the equator did not influence the forecast in the mid-latitudes for nearly too weeks. However, walls north of the equator damaged the results after less than a week. Different boundary conditions have little effect on the forecast except near regions of high wind speed.

Full access
DAVID P. BAUMHEFNER

Abstract

A number of global real-data numerical forecasts have been calculated using the two-layer NCAR (National Center for Atmospheric Research) general circulation model. The purpose of these experiments was threefold: 1) to evaluate the model's ability to predict the real atmosphere, 2) to develop a global forecasting model which will make use of the data obtained by the proposed GARP (Global Atmospheric Research Program), and 3) to help determine some of the internal, empirical constants of the model. In order to evaluate the accuracy of the predictions, several “skill scores” were calculated from the forecasted and observed variables. A by-product of this research was the testing of five different types of data-initialization schemes. Over 50, 4-day forecasts have been run, in which the initialization schemes and internal constants were varied.

The results from these experiments indicate that the present two-layer model is capable of forecasting the real atmosphere with reasonable skill out to 2 days at the surface and 4 days in the middle troposphere. The best initialization scheme for this particular model, thus far, appears to be the complete balance equation. However, several of the simplified initialization techniques are very close in terms of forecasting skill.

Full access
DAVID P. BAUMHEFNER

Abstract

A diagnostic, nonlinear balanced model is applied in order to describe numerically the three dimensional structure of the tropical atmosphere. Several comparisons and experiments are made to gain insight into the physical processes and reliability of the model. These include different types of stream functions and temperature analyses, and the addition of surface friction and latent heat. A comparison between the kinematic vertical motion and the final numerical result is performed.

Obtained by using the complete form of the balance model, the derived vertical motion for August 12–14, 1961, in the Caribbean is presented in the form of cross sections. The vertical velocity fields, which are displayed in partitioned form, are compared with the analyzed moisture distribution. The validity of the computed vertical motion is discussed along with its possible influence on the tropical weather.

Full access
Stephen J. Colucci
and
David P. Baumhefner

Abstract

The forecastability of a blocking episode during January 1985 over Europe and the eastern Atlantic Ocean is studied with forecast ensembles. Ten-member ensembles from version CCM2 (at T42 resolution) of the Community Climate Model of the National Center for Atmospheric Research are initialized at various lead times prior to the analyzed block onset and run out to 14 days. Particular attention is focused on the ensemble initialized five days prior to block onset since, of all the ensembles, this one was characterized by the greatest variability concerning the block-onset prediction. Two of the 10 members of this particular ensemble predicted a transition from unblocked to blocked flow over the Atlantic–Europe half of the Northern Hemisphere during the 14-day forecast range, but not without error; details regarding the timing and/or location of the block were misforecast. A comparison of these ensemble members, plus one other that did not predict a transition to blocking, with the corresponding analyses revealed that the block forecast errors could be traced to a model failure to predict the anomalously weakened midtropospheric planetary-scale geostrophic westerlies analyzed upstream of and prior to the block onset. The forecast error appeared to be attributable to a model bias toward an erroneously southward displacement of the midtropospheric zonal jet over Europe and the eastern Atlantic Ocean. On the other hand, the interaction between the planetary-scale flow and synoptic-scale activity, as measured by the midtropospheric advection of synoptic-scale quasigeostrophic potential vorticity by the planetary-scale geostrophic wind, was well predicted by the ensemble members, but perhaps fortuitously. The results demonstrate that the forecast ensemble was able to overcome the influence of the systematic error by indicating the possibility of a transition to blocked flow over the domain and within the forecast range.

Full access
Steven L. Mullen
and
David P. Baumhefner

Abstract

The impact of initial condition uncertainty on short-range (0–48 h) simulations of explosive surface cyclogenesis is examined within the context of a perfect model environment. Eleven Monte Carlo simulations are performed on 10 cases of rapid oceanic cyclogenesis that occurred in a long-term, perpetual January integration of a global spectral model. The perturbations used to represent the initial condition error have a magnitude and spatial decomposition that closely matches estimates of global analysis error.

Large variability characterizes the error growth rates, both among the individual Monte Carlo simulations and among the case-average values. Some individual simulations display error growth doubling times as fast as approximately 12 h during the 24-h period of most rapid intensification, while others exhibit virtually no error growth. The variability is also reflected in the wide 90% confidence bounds for many surface weather elements such as the cyclone position and central pressure. However, no statistically significant differences are found between the initial states leading to large simulation errors and those leading to negligible errors. These results attest to the importance of initial condition uncertainty as the major cause of forecast variability and indicate a strong sensitivity to subtle differences in initial perturbation location and structure.

The effect that simple ensemble averaging has on reducing uncertainty is discussed. Averaging a 16-member ensemble decreases the random component of the initial data error by 80%–90% and the 90% confidence bounds by 70%–80% for cyclone position, central pressure, and 12-h pressure change. It is hypothesized that ensemble forecasting could benefit the utility of short-range forecasts for many weather elements of operational interest and conclude that research efforts should be directed at examining its effectiveness in an operational setting.

Full access
Warren M. Washington
and
David P. Baumhefner

Abstract

A simple method of reducing the amplitude of Lamb waves in primitive equation model forecasts has been proposed and tested. This method makes use of a Boussinesq-type approximation in which the vertical mean mass divergence is set equal to zero. It effectively reduces the Lamb waves by a factor of 3 in the example shown here and does not degrade the forecast accuracy. The largest reduction in Lamb wave amplitude is found in the tropical regions.

Full access
David P. Baumhefner
and
Paul R. Julian

Abstract

A reference level is defined as a level of known altitude at which temperature, pressure, and perhaps wind are specified as functions of time. This study is concerned with the optimum location of a reference level without wind information. Experiments were performed with the NCAR six-layer general circulation model to compare the usefulness of a surface reference level with an upper tropospheric reference level. We first performed a control integration using real atmospheric initial data. We then ran several comparison runs with initial conditions differing from those of the control run. The initial pressure distribution at the reference level was kept the same as the control run. The distribution of temperature pseudo-error employed in calculating the initial pressure distributions at the other levels was chosen to simulate possible error patterns in temperatures radiometrically derived from satellites. The initial conditions in all cases were in hydrostatic and geostrophic balance. Three data sets were used and the experiments were integrated to five or seven days. In addition, two horizontal distributions of initial temperature pseudo-error and two horizontal mesh lengths of the model were used for one of the three data sets. The results were examined using an rms difference of the distribution of pressure and meridional wind normalized (in the vertical) by the difference statistics derived from randomly chosen states of the model.

It appears that pseudo-error growth rates are nearly independent of the location of a reference level, but details of the pseudo-error patterns depend on the initial synoptic conditions. Pseudo-error growth rates differed depending on the manner in which the horizontal pseudo-error was initially distributed (but did not differ with the location of the reference level). The most significant change in the pseudo-error growth rates was observed when the mesh length was changed; halving the mesh length produced much faster growth rates, particularly in the lower layers.

Full access
Joseph J. Tribbia
and
David P. Baumhefner

Abstract

This paper presents the results of an ensemble of 20 predictability experiments derived from the NCAR Community Climate Model (CCM). Particular emphasis is placed on the question of the predictability of dynamically driven low-frequency components of the model atmosphere. The conclusion drawn, using time averaging alone as a means of isolating low-frequency variability, is that in the ensemble mean there is little skill in a 30-day mean forecast. Examination of the variability of skill among the ensemble members indicates that approximately 40 percent of the perturbed monthly mean forecasts would be useful. Examples of skillful and poor monthly mean predictions am shown and conclusions are drawn as to the implications of the results with regard to the likelihood of success of extended range deterministic forecasts.

Full access
Stephen J. Colucci
and
David P. Baumhefner

Abstract

A set of thirty 30-day mean 500-mb height anomaly forecasts run from National Meteorological Center initial analyses by the NCAR Community Climate Model is examined in order to learn if the forecast accuracy can be estimated with the initial conditions. Defining initial weather regimes by a 500-mb geostrophic zonal index anomaly difference between 50°W and 10°E discriminates between the best and worst 30-day mean forecasts in the sample. Initial regimes characterized by anomalously high zonal index (500-mb geostrophic westerlies) at 50°W and low index at 10°E yield on average lower 30-day mean forecast-observed anomaly correlation than initial regimes with opposite conditions (anomalously low zonal index at 50°N and high index at 10°E). It is suggested that initial regimes with abnormally fast geostrophic 500-mb westerlies at 50°W are followed in time by intense and poorly forecast synoptic-scale cyclones over the Atlantic Ocean. It is shown in a case study that the local synoptic- to planetary-scale interaction, as measured by its contribution to quasigeo-strophic 500-mb height tendencies, is misforecast early in the forecast cycle following these initial conditions. Early rapid synoptic-scale error growth, in this case, is followed by rapid planetary-scale error growth (incorrectly forecast demise of a blocking pattern), deterioration of forecast accuracy, and an unskillful 30-day mean forecast. It is further suggested that initial regimes with abnormally weak 500-mb geostrophic westerlies at 50°W are followed in time by less intense, but better forecast, cyclone waves over the Atlantic Ocean. In a representative case study, the local synoptic- to planetary-scale interaction is well forecast early in the forecast cycle. The local synoptic- and planetary-scale error growths are more restrained (a blocking pattern is correctly maintained), deterioration of forecast accuracy is postponed, and a skillful 30-day mean forecast results. It is hypothesized on the basis of this work that the accuracy of numerical 30-day mean forecasts may depend upon the accuracy with which the cyclones and their interactions with the planetary scale are predicted early in the forecast cycle, and that this accuracy in turn may depend upon the initial conditions.

Full access